Re: EXT2-fs error and geometry walk ... ???
Andries, The crap out is between 2.4.0-test5 and 2.4.0-test6. It takes four drives that were single partitioned and rips the first 130 blocks out and creates 4 bogus partitions. Partition check: /dev/ide/host0/bus0/target0/lun0: p1 p2 p3 < p5 p6 p7 p8 p9 p10 > /dev/ide/host0/bus0/target1/lun0: p1 /dev/ide/host2/bus0/target0/lun0: p1 p2 p3 p4 /dev/ide/host2/bus0/target1/lun0: p1 p2 p3 p4 /dev/ide/host2/bus0/target2/lun0: p1 p2 p3 p4 /dev/ide/host2/bus0/target3/lun0: p1 p2 p3 p4 Was Partition check: /dev/ide/host0/bus0/target0/lun0: p1 p2 p3 < p5 p6 p7 p8 p9 p10 > /dev/ide/host0/bus0/target1/lun0: p1 /dev/ide/host2/bus0/target0/lun0: p1 /dev/ide/host2/bus0/target1/lun0: p1 /dev/ide/host2/bus0/target2/lun0: p1 /dev/ide/host2/bus0/target3/lun0: p1 This is the difference between boots. Cheers, Andre Hedrick The Linux ATA/IDE guy - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: EXT2-fs error and geometry walk ... ???
On Mon, 18 Sep 2000, Andries Brouwer wrote: > If this is an unpatched vanilla 2.4.0test8 then I am surprised. > But if it is a patched version I would prefer to blame the patch. Andries, Going back to 2.4.0test5 with my patch it works perfectly. I will move this forward to 2.4.0test8, but all the patch does, is to expand the select method beyond the SELECT_DRIVE(hwif,drive). Since all drives report as an effective master, the select code only raise the device to that level. Thus I tend to focus on filesystems. The fact that partition check points are were it craps out on two different systems. One Cascade Project and the other is the DupliDisk II RaidCase, by sratching the first superblock on every other boot and not in 2.2.X kernels or 2.4.0-test5 currently testing and stepping forward. Cheers, Andre Hedrick The Linux ATA/IDE guy - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: EXT2-fs error and geometry walk ... ???
On Mon, Sep 18, 2000 at 04:50:55AM -0700, Andre Hedrick wrote: [ext2 errors and fdisk complaints on 2.4.0test8, patched?] Andre, (i) Geometry does not play any role in the functioning of Linux - it is only a matter to LILO and fdisk. So, if you meet a strange geometry, then that is surprising, but cannot cause other errors. Of course it is possible that the same bug causes both ext2 errors and a strange geometry report. (ii) You find exactly the same fdisk oddities, and exactly the same ext2 errors on each of four disks. This sounds more like a programming than like a hardware error, although of course I do not know what kind of hardware setup you use. So, perhaps you should explain what you do. If this is an unpatched vanilla 2.4.0test8 then I am surprised. But if it is a patched version I would prefer to blame the patch. > ext2_readdir: bad entry in directory #2: rec_len % 4 != 0 - offset=0, inode=2, >rec_len=14, name_len=1 > > This did not happen in the past. Do you have a reproducible situation, with a kernel that functions and a kernel that fails? Andries - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: EXT2-fs error and geometry walk ... ???
Linux cascade 2.4.0-test8 #1 Sat Sep 16 23:50:47 PDT 2000 i686 unknown Andre Hedrick The Linux ATA/IDE guy - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: EXT2-fs error and geometry walk ... ???
On Mon, 18 Sep 2000, Andre Hedrick wrote: > On Mon, 18 Sep 2000, Alexander Viro wrote: > > > > > > > Erm. What version it was? > > 1.18-125 (installed 1.18-125) SuSE 7.0 Professional > > If you want the srpm will send it offline. Sorry, what kernel version were you using? - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: EXT2-fs error and geometry walk ... ???
On Mon, 18 Sep 2000, Alexander Viro wrote: > > > Erm. What version it was? 1.18-125 (installed 1.18-125) SuSE 7.0 Professional If you want the srpm will send it offline. Andre Hedrick The Linux ATA/IDE guy - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: EXT2-fs error and geometry walk ... ???
Erm. What version it was? - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
EXT2-fs error and geometry walk ... ???
Please not that all of this is the same boot. Disregard the weird device major:minor Sep 18 04:22:20 cascade kernel: EXT2-fs error (device ide2(33,1)): ext2_readdir: bad entry in directory #2: rec_len % 4 != 0 - offset=0, inode=2, rec_len=14, name_len=1 Sep 18 04:22:20 cascade kernel: EXT2-fs error (device ide2(33,17)): ext2_readdir: bad entry in directory #2: rec_len % 4 != 0 - offset=0, inode=2, rec_len=14, name_len=1 Sep 18 04:22:20 cascade kernel: EXT2-fs error (device ide2(33,33)): ext2_readdir: bad entry in directory #2: rec_len % 4 != 0 - offset=0, inode=2, rec_len=14, name_len=1 Sep 18 04:22:20 cascade kernel: EXT2-fs error (device ide2(33,49)): ext2_readdir: bad entry in directory #2: rec_len % 4 != 0 - offset=0, inode=2, rec_len=14, name_len=1 This did not happen in the past. cat /proc/partitions 33 0 15087744 ide/host2/bus0/target0/lun0/disc 33 1 15085003 ide/host2/bus0/target0/lun0/part1 33167543872 ide/host2/bus0/target1/lun0/disc 33177542486 ide/host2/bus0/target1/lun0/part1 33327543872 ide/host2/bus0/target2/lun0/disc 33337542486 ide/host2/bus0/target2/lun0/part1 3348 15087744 ide/host2/bus0/target3/lun0/disc 3349 15085003 ide/host2/bus0/target3/lun0/part1 cat /proc/ide/hde*/geometry physical 29936/16/63 logical 1878/255/63 physical 14968/16/63 logical 939/255/63 physical 14968/16/63 logical 939/255/63 physical 29936/16/63 logical 1878/255/63 I need to know what is going on in ext2 that now makes this error. What is causing the geometry walk? Disk /dev/hdea: 255 heads, 63 sectors, 1878 cylinders Units = cylinders of 16065 * 512 bytes Device BootStart EndBlocks Id System /dev/hdea1 9 1895 15150539+ 83 Linux Disk /dev/hdeb: 255 heads, 63 sectors, 939 cylinders Units = cylinders of 16065 * 512 bytes Device BootStart EndBlocks Id System /dev/hdeb1 9 948 7542486 83 Linux Partition 1 has different physical/logical beginnings (non-Linux?): phys=(0, 1, 3) logical=(8, 41, 33) Partition 1 has different physical/logical endings: phys=(938, 254, 63) logical=(947, 40, 32) Disk /dev/hdec: 255 heads, 63 sectors, 939 cylinders Units = cylinders of 16065 * 512 bytes Device BootStart EndBlocks Id System /dev/hdec1 9 948 7542486 83 Linux Partition 1 has different physical/logical beginnings (non-Linux?): phys=(0, 1, 3) logical=(8, 41, 33) Partition 1 has different physical/logical endings: phys=(938, 254, 63) logical=(947, 40, 32) Disk /dev/hded: 255 heads, 63 sectors, 1878 cylinders Units = cylinders of 16065 * 512 bytes Device BootStart EndBlocks Id System /dev/hded1 9 1895 15150539+ 83 Linux Sep 18 04:36:21 cascade kernel: /dev/ide/host2/bus0/target3/lun0: p1 p2 p3 p4 Sep 18 04:37:04 cascade kernel: /dev/ide/host2/bus0/target2/lun0: p1 p2 p3 p4 Sep 18 04:37:06 cascade kernel: /dev/ide/host2/bus0/target2/lun0: p1 p2 p3 p4 Sep 18 04:37:22 cascade kernel: /dev/ide/host2/bus0/target1/lun0: p1 p2 p3 p4 Sep 18 04:41:19 cascade kernel: /dev/ide/host2/bus0/target0/lun0: p1 p2 p3 p4 Expert command (m for help): p Disk /dev/hdea: 255 heads, 63 sectors, 1878 cylinders Nr AF Hd Sec Cyl Hd Sec CylStart Size ID 1 00 0 00 0 0000 00 2 00 0 20 0 20 131072 131072 00 3 00 0 20 0 20 131072 131072 00 4 00 0 20 0 20 131072 131072 00 out of expert mode and delete all 4 bogus partitions. back in expert mode Expert command (m for help): p Disk /dev/hdea: 255 heads, 63 sectors, 1878 cylinders Nr AF Hd Sec Cyl Hd Sec CylStart Size ID 1 00 0 00 0 0000 00 2 00 0 00 0 0000 00 3 00 0 00 0 0000 00 4 00 0 00 0 0000 00 cat /proc/partitions 33 0 15087744 ide/host2/bus0/target0/lun0/disc 33 1 65536 ide/host2/bus0/target0/lun0/part1 33 2 65536 ide/host2/bus0/target0/lun0/part2 33 3 65536 ide/host2/bus0/target0/lun0/part3 33 4 65536 ide/host2/bus0/target0/lun0/part4 33167543872 ide/host2/bus0/target1/lun0/disc 3317 65536 ide/host2/bus0/target1/lun0/part1 3318 65536 ide/host2/bus0/target1/lun0/part2 3319 65536 ide/host2/bus0/target1/lun0/part3 3320 65536 ide/host2/bus0/target1/lun0/part4 33327543872 ide/host2/bus0/target2/lun0/disc 3333 65536 ide/host2/bus0/target2/lun0/part1 3334 65536 ide/host2/bus0/target2/lun0/part2 3335 65536 ide/host2/bus0/target2/lun0/part3 3336 65536 ide/host2/bus0/target2/lun0/part4 3348 15087744 ide/host2/bus0/target3/lun0/disc 3349 65536 ide/host2/bus0/target3/lun0/p
